fix(frontend): add support for 'mode' discriminator in oneOf rendering

Update ArgInput.svelte to properly handle oneOf schemas that use 'mode' as the
discriminator field, in addition to the existing 'kind' and 'label' support.

Changes:
- Updated tagKey derivation to check for 'mode' first, then 'kind', then 'label'
- Added 'mode' to the onOneOfChange function to track mode changes
- Added 'mode' to the list of keys excluded from enum validation
- Added 'mode' to hiddenArgs to prevent it from being shown in the form
- Added title fields to the history oneOf variants in flowInfers.ts

This allows the AI agent's history field to properly render with toggle buttons
for 'auto' and 'manual' modes.

Co-authored-by: centdix <centdix@users.noreply.github.com>

Author: github-actions[bot]

backend/windmill-worker/src/ai_executor.rs

@@ -424,14 +424,14 @@ pub async fn run_agent(
 
     // Load messages based on history mode
     if matches!(output_type, OutputType::Text) {
-        match history {
+        match &history {
             Some(History::Manual { messages: manual_messages }) => {
                 // Use explicitly provided messages (bypass memory)
                 if !manual_messages.is_empty() {
-                    messages.extend(manual_messages);
+                    messages.extend(manual_messages.clone());
                 }
             }
-            Some(History::Auto { context_length }) if context_length > 0 => {
+            Some(History::Auto { context_length }) if *context_length > 0 => {
                 // Auto mode: load from memory
                 if let Some(step_id) = job.flow_step_id.as_deref() {
                     if let Some(memory_id) = flow_context
@@ -443,7 +443,7 @@ pub async fn run_agent(
                         match read_from_memory(db, &job.workspace_id, memory_id, step_id).await {
                             Ok(Some(loaded_messages)) => {
                                 // Take the last n messages
-                                let start_idx = loaded_messages.len().saturating_sub(context_length);
+                                let start_idx = loaded_messages.len().saturating_sub(*context_length);
                                 let mut messages_to_load = loaded_messages[start_idx..].to_vec();
                                 let first_non_tool_message_index =
                                     messages_to_load.iter().position(|m| m.role != "tool");
@@ -934,16 +934,16 @@ pub async fn run_agent(
     // Skip memory persistence if using manual messages (bypass memory entirely)
     // final_messages contains the complete history (old messages + new ones)
     if matches!(output_type, OutputType::Text) && !use_manual_messages {
-        if let Some(History::Auto { context_length }) = history {
-            if context_length > 0 {
+        if let Some(History::Auto { context_length }) = &history {
+            if *context_length > 0 {
                 if let Some(step_id) = job.flow_step_id.as_deref() {
                     // Extract OpenAIMessages from final_messages
                     let all_messages: Vec<OpenAIMessage> =
                         final_messages.iter().map(|m| m.message.clone()).collect();
 
                     if !all_messages.is_empty() {
                         // Keep only the last n messages
-                        let start_idx = all_messages.len().saturating_sub(context_length);
+                        let start_idx = all_messages.len().saturating_sub(*context_length);
                         let messages_to_persist = all_messages[start_idx..].to_vec();
 
                         if let Some(memory_id) = flow_context.flow_status.and_then(|fs| fs.memory_id) {

frontend/src/lib/components/ArgInput.svelte

@@ -202,7 +202,11 @@
 
 	let oneOfSelected: string | undefined = $state(undefined)
 	let tagKey = $derived(
-		oneOf?.find((o) => Object.keys(o.properties ?? {})?.includes('kind')) ? 'kind' : 'label'
+		oneOf?.find((o) => Object.keys(o.properties ?? {})?.includes('mode'))
+			? 'mode'
+			: oneOf?.find((o) => Object.keys(o.properties ?? {})?.includes('kind'))
+				? 'kind'
+				: 'label'
 	)
 	async function updateOneOfSelected(oneOf: SchemaProperty[] | undefined) {
 		if (
@@ -226,10 +230,13 @@
 	function onOneOfChange() {
 		const label = value?.['label']
 		const kind = value?.['kind']
+		const mode = value?.['mode']
 		if (label && oneOf && oneOf.some((o) => o.title == label) && oneOfSelected != label) {
 			oneOfSelected = label
 		} else if (kind && oneOf && oneOf.some((o) => o.title == kind) && oneOfSelected != kind) {
 			oneOfSelected = kind
+		} else if (mode && oneOf && oneOf.some((o) => o.title == mode) && oneOfSelected != mode) {
+			oneOfSelected = mode
 		}
 	}
 
@@ -1073,7 +1080,7 @@
 								for (const key of newValueKeys) {
 									// Check if there is a select (enum) in the newly selected oneOf and if the current value is not in the enum, skip it
 									if (
-										!['kind', 'label'].includes(key) &&
+										!['kind', 'label', 'mode'].includes(key) &&
 										selectedObjProperties[key]?.enum &&
 										value &&
 										value[key] !== undefined &&
@@ -1083,9 +1090,11 @@
 									}
 									toKeep[key] = value[key]
 								}
-								const tagKey = oneOf.find((o) => Object.keys(o.properties ?? {}).includes('kind'))
-									? 'kind'
-									: 'label'
+								const tagKey = oneOf.find((o) => Object.keys(o.properties ?? {}).includes('mode'))
+									? 'mode'
+									: oneOf.find((o) => Object.keys(o.properties ?? {}).includes('kind'))
+										? 'kind'
+										: 'label'
 								value = { ...toKeep, [tagKey]: detail }
 							}}
 						>
@@ -1122,9 +1131,11 @@
 											}
 											bind:args={value}
 											hiddenArgs={[
-												oneOf?.find((o) => Object.keys(o.properties ?? {}).includes('kind'))
-													? 'kind'
-													: 'label'
+												oneOf?.find((o) => Object.keys(o.properties ?? {}).includes('mode'))
+													? 'mode'
+													: oneOf?.find((o) => Object.keys(o.properties ?? {}).includes('kind'))
+														? 'kind'
+														: 'label'
 											]}
 											on:reorder={(e) => {
 												if (oneOf && oneOf[objIdx]) {
@@ -1143,7 +1154,7 @@
 											{disablePortal}
 											{disabled}
 											{prettifyHeader}
-											hiddenArgs={['label', 'kind']}
+											hiddenArgs={['label', 'kind', 'mode']}
 											schema={{
 												properties: obj.properties,
 												order: obj.order,

frontend/src/lib/components/flows/flowInfers.ts

@@ -39,7 +39,8 @@ export const AI_AGENT_SCHEMA = {
 				'Configure how conversation history is managed. Choose "auto" to automatically store and load messages from memory (up to N last messages), or "manual" to provide an explicit array of conversation messages. When using manual mode, memory is bypassed entirely - messages are not loaded from or saved to memory. The system_prompt (if provided) is always prepended, and user_message (if provided) is always appended as the final message.',
 			oneOf: [
 				{
-					type: 'object' as const,
+					type: 'object',
+					title: 'auto',
 					properties: {
 						mode: {
 							type: 'string',
@@ -59,7 +60,8 @@ export const AI_AGENT_SCHEMA = {
 						'When no S3 storage is configured in your workspace settings, memory will be stored in database, which implies a limit of 100KB per memory entry. If you need to store more messages, you should use S3 storage in your workspace settings.'
 				},
 				{
-					type: 'object' as const,
+					type: 'object',
+					title: 'manual',
 					properties: {
 						mode: {
 							type: 'string',
@@ -71,7 +73,7 @@ export const AI_AGENT_SCHEMA = {
 							type: 'array',
 							description: 'Array of conversation messages to use as history',
 							items: {
-								type: 'object' as const,
+								type: 'object',
 								properties: {
 									role: {
 										type: 'string',
@@ -84,12 +86,12 @@ export const AI_AGENT_SCHEMA = {
 										type: 'array',
 										nullable: true,
 										items: {
-											type: 'object' as const,
+											type: 'object',
 											properties: {
 												id: { type: 'string' },
 												type: { type: 'string' },
 												function: {
-													type: 'object' as const,
+													type: 'object',
 													properties: {
 														name: { type: 'string' },
 														arguments: { type: 'string' }
@@ -112,7 +114,7 @@ export const AI_AGENT_SCHEMA = {
 				}
 			],
 			showExpr: "fields.output_type === 'text'"
-		},
+		} as any,
 		output_schema: {
 			type: 'object',
 			description: 'JSON schema that the AI agent will follow for its response format.',